This paper describes a two-step Viterbi decoding based on reinforcement learning and information theory with telephone speech. The idea is to strength or weaken HMM's by using Bayes-based confidence measure (BBCM) and distances between models. If HMM's in the N-best list show a low BBCM, the second Viterbi decoding will prioritize the search on neighboring models according to their distances to the N-best HMM's. The current reinforcement learning mechanism is modeled as the linear combination of two metrics or information sources. Moreover, a criterion based on incremental conditional entropy maximization to optimize a linear combination of metrics or information sources is also presented. As shown here, the method requires only one adapting utterance and can lead to a reduction in WER as high as 10.9%.
Bibliographic reference. Molina, Carlos / Yoma, Nestor Becerra / Huenupan, Fernando / Garreton, Claudio (2008): "Unsupervised re-scoring of observation probability based on maximum entropy criterion by using confidence measure with telephone speech", In INTERSPEECH-2008, 1016-1019.